Currently available evoked potential machines generally rely on ensemble averaging to separate the evoked potential waveforms from accompanying noise. This technology imposes limits on the utility of these systems in "high-noise" environments, and when rapid resolution of potentially changable evoked potentials in clinically essential, such the in operating room and intensive care unit. Corteks proposes the development of an evaoked potential machine which will resolve evoke potential waveforms more rapidly and with substantially less noise contamination than currently available commercial systems. This device's advantages will be derived from the use of a combination of digital signal processing techniques includng phase-synchonized triggering, two- dimensional filtering, statistical outlier elimination, and high- speed pseudorandom stimulation. In Phase I we will implement and critically evalutate a real-time prototype. In Phase II we anticipate replacing the array processor used in the Phase I prototype with a superior and lower cost device, implement high-speed pseudorandom stimulation, as well as automatic optimization algorithms to simplify use.